e-water · circulating within the calcarenitic aquifer. however, the lake waters chemical...

E-WAter Official Publication of the European Water Association (EWA) © EWA 2006

- 1 -

Cusimano G. 1,3, Hauser S. 2,3 *, Vassallo M.2

Hydrogeochemistry of a wetland area of southwestern Sicily (Italy) ABSTRACT The “Preola” and “Gorghi tondi” lakes are the most noticeable wetlands in the Mazara del Vallo territory (south-west Sicily). There are four karst origin lakes located in a natural depression formed by gypsum dissolution and subsequent collapse of the “Calcarenite di Marsala”, a few meters above sea level. Erosion of the depression sides and human activity has caused visible colluvial deposits, which have contributed to a natural filling process of the lakes. A hydrogeological map of the area and the reconstruction of a water circulation model supplying the lakes have been drafted from geochemical and hydrogeological data. It is apparent from δ18O values that the well waters are mainly fed by local meteoric water circulating within the calcarenitic aquifer. However, the lake waters chemical composition are very likely controlled by evaporation processes and a complex mixing between seawater and groundwater, depending on seasonal variation in the hydrodynamic equilibrium. SOMMARIO I laghi “Preola e Gorghi tondi” sono le aree umide più importanti del territorio di Mazara del Vallo (Sicilia sud-occidentale). Si tratta di quattro laghi di origine carsica situati in una depressione originata dalla dissoluzione di gessi e conseguente crollo della sovrastante “Calcarenite di Marsala”, ad una quota prossima al livello del mare. L’erosione e le attività umane sono la causa della formazione di depositi colluviali che hanno contribuito al naturale processo di interramento dei laghi. Con l’acquisizione di dati idrogeologici e geochimici è stato possibile disegnare una carta idrogeologica, linee di deflusso sotterraneo un modello di circolazione delle acque. I dati isotopici del δ18O dell’acqua di falda confermano che l’alimentazione dell’acquifero calcarenitico è dovuta ad acque meteoriche locali. Tuttavia la composizione chimica dei laghi e controllata da processi di evaporazione e di miscela tra acqua di mare e di falda, sensibilmente dipendenti dalle variazioni dell’equilibrio idrodinamico. Keywords: Hydrogeochemistry, Hydrogeology, Isotopes, Karst.

1 Dipartimento di Geologia e Geodesia - Università di Palermo 2 Dipartimento di Chimica e Fisica della Terra (CFTA) - Università di Palermo, Via Archirafi 36, 90123 Palermo 3 Centro Interdipartimentale Ricerche Interazione Tecnologia Ambiente - Università di Palermo * Corresponding author, e-mail: [email protected], FAX: 0039 091 6168376


- 2 -

INTRODUCTION About two thirds of the wetland areas in the Mediterranean basin have been drained, while most of the remaining area continues to deteriorate. Aquifer overexploitation in coastal areas is a serious problem. Many coastal aquifer systems in Italy have been so seriously depleted that their piezometric surface has been lowered, causing progressive seawater intrusion [17]. The restoration of Mediterranean wetlands has become a priority. Every Mediterranean country has expressed, through participation in the Mediterranean Wetlands Initiative [6,9,19], their decision to promote studies implementing the knowledge of the delicate equilibrium controlling these areas and restoration projects. Moreover, the evaporate sequence of the Upper Miocene is present in Italy in several regions such as Tuscany, Calabria and Sicily and many studies regarding karst processes in the evaporate sequences are also reported elsewhere [7,10]. The area investigated, located in the middle of the Mediterranean basin, is of relevant interest for both themes. The lakes “Preola” (named in the text as “LP”) and “Gorghi tondi” (as “G1”, “G2”, and “G3”) are located in southwestern Sicily, near “Mazara del Vallo” town (Figure 1).

Figure 1: Location of the study area.

Today this area is a Natural Reserve managed by the WWF (World Wide Fund for Nature) Italy and represents a unique ecosystem for its great variety of fauna that flourishes in this particular wetland microclimate. At the moment, one of the most important management problems of the Natural Reserve is to restore the hydrological equilibrium in the lakes. In fact, this wetland is an ideal rest site for migratory birds, as Gray Heron (Ardea cinerea), Little Egret (Egretta garzetta), Black-winged Stilt (Himantopus himantopus), Eurasian Coot (Fulica atra), Common Moorhen (Gallinula chloropus) and other species from North Africa to central Europe.


- 3 -

Figure 2 shows the location of the field sites. It is delimited to the north by the “Palermo-Mazara del Vallo” freeway, and to the south and west by the Mediterranean Sea.

Figure 2: Hydrogeological map and sampling sites


- 4 -

The landscape morphology is essentially flat, with a slight incline towards the sea. This trend is interrupted by a karst depression, created by the subsiding calcarenitic cover. The depression is about 20 m deep and runs parallel to the coastal line for about 4 km, at an average distance of 1.7 km, marked by the presence of lakes and dolines. For several years, the investigated aquifer, located in the extensive quaternary coastal plain, has shown signs of continuous impoverishment as a consequence of uncontrolled water exploitation. This has caused a subsequent decrease in groundwater contribution to the lakes [16]. The drastic drawdown of the hydrostatic level and modification of the delicate natural equilibrium has caused the disappearance of some surrounding wet zones (such as the nearby lake “Murana”). The research has been focused on:

� The definition of underground lithology related to the karst lakes; � Piezometric surface reconstruction in order to define groundwater circulation and the

relationships between the aquifer, the lakes and the sea; � Aquifer characterization and its probable recharge area, using geochemical and

isotopic data.

METHODS Well water samples were collected in three different periods: July 2000, April and October 2001. Seawater samples were also analysed. Meteoric samples were collected monthly for a period of two years by a rain gage installed within the study area. During the sample collection, some parameters were measured in situ: Temperature (thermocouple, ± 0,1 °C), pH and Eh (WTW pH 315, ± 0,01mV), electrical conductivity (WTW Cond. 315, mS/cm at 20°C, ± 0,5% of the measured value). The chemical-physical parameters in the lake waters were measured by a multi-parametric probe (Amel 396) at different points and depths to check the possible existence of a vertical gradient in the water column. Sample aliquots, collected in polyethylene bottles, were filtered through 0.45 µm Millipore filter and acidified by HNO3 ultra-pure at pH 2. Alkalinity was measured by HCl titration, on un-acidified aliquot, and chemical constituents by ion chromatography (Dionex Dx 1-120; Li+, F-, Br-;error was calculated around 3% for high concentration and 7% for low concentration; SO4=,Cl-, NO3-, Na+, K+ Mg2+, Ca2+ was 2% for high and for 5% for low concentration, respectively). Sampling techniques and chemical analyses have been carried out according to: “Standard Methods for examination of water and wastewater” [13]. δ18O measurements were performed, on an unfiltered sample, using the CO2-H2O equilibration procedure reported the Epstein and Mayeda [8]. The reproducibility was ± 0,2 ‰. The δD analyses were carried out using the H2O-Zn reduction [4]. The reproducibility was ± 1‰. Isotopic analyses have been reported against the International Standard V-SMOW as defined by R.Gonfiantini [12].


- 5 -

RESULTS AND DISCUSSION HydrogeologyThe area, within the neogenic basin of “Castelvetrano”, is constituted by upper Miocene and Tyrrhenian terrains [1,2,3]. The Pleistocene calcarenitic successions constitute the most extensive hydrogeological unit in the zone and contain an unconfined aquifer, locally in hydraulic contact with the underlying aquifer (evaporitic complex). The main hydrogeological units are shown in Figure 2. The hydrostatic level was reconstructed through piezometric wells measurements and hydric levels of the lakes. The measurements were carried out in a short period (one-week) and during periods when the wells were not being exploited for agricultural purposes. Interpretation of the piezometric contours has revealed:

� a parallel trend of piezometric lines and the karst depression, demonstrating a water supply towards the lakes;

� a hydrogeological boundary in the sectors where the calcarenitic aquifer unconforms the clayey-sandy clayey substratum.

HydrogeochemistryThe Piper [15] diagram was preferred to other conventional methods because it allows a more precise identification of the water samples and some dominant geochemical processes in the water chemistry (Fig. 3).

Figure 3: Piper plot showing major chemical compositions of the ground waters and lakes The diagram also shows the seawater point (SW). At first glance one can see the presence of two main water groups: one of Na+, K+ - Cl-, SO4= composition (essentially lake waters and well P12), and the other of Ca2+, Mg2+ - Cl-, SO4= type (the remaining wells). Considering the


- 6 -

ternary diagrams, it can be concluded that the water chemistry is mainly controlled by Na+-K+- Ca2+ cations and Cl- - SO4=, and subordinately HCO3- anions. This first analysis reveals that seawater, the calcarenitic aquifer, and the presence of gypsum control the water chemistry. For instance, the Cl- iso-concentration lines (Figure 4) point out the possible seawater intrusion direction.

Figure 4: Cl- iso-concentration map (concentration expressed in meq/l) The Cl- - Na+ diagram (Figure 5) shows that, at low concentration, almost all samples result to be aligned along a seawater mixing line, except the lake LP points which depart from this line for all the samplings. Considering that during drought periods (Summer time) LP dries completely (to collect water samples it was necessary to drill at least 2 m beneath the lake bottom), it seems reasonable to think that last stage mineralogical phases of the evaporitic series (XRD determination performed on lake sediments showed peaks related to Kainite and Caminite) can be precipitated. When water once again reaches the lake, there is redissolution of these phases.


- 7 -

LP july 2000

LP apr. 2001

LP oct. 2001

Seawater

0100200300400500600700

0 100 200 300 400 500 600 700Na (meq/l)

Cl(m

eq/l)

Figure 5: Cl- vs. Na+ plot Even Figure 6 and Figure 7 (the Cl- vs. Mg2+ and SO4= contents) show that these two ions, which are second in abundance in the seawater after Na+ and Cl-, don’t follow the seawater mixing line. Whereas the Cl-/Na+ ratio is close to 1.0 in the seawater (1.2 in the LP lake), the Cl-/Mg2+ (33.5) and Cl-/SO4= (9.2) ones are completely different with respect to those of LP lake (3.2, 2.5, 3.8 and 3.1, 0.9, 5.6, respectively for July 2000, April 2001 and October 2001). This means that large changes have occurred in the water chemistry of LP lake with respect to the seawater, very likely due, as previously said, to precipitation of some mineralogical phases.

LP july 2000

LP apr. 2001

LP oct. 2001

Seawater

0100200300400500600700

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150Mg (meq/l)

Cl(m

eq/l)

Figure 6: Cl- vs. Mg2+ plot

LP july 2000

LP apr. 2001

LP oct. 2001

Seawater

0100200300400500600700

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150SO4 (meq/l)

Cl(m

eq/l)

Figure 7: Cl- vs. SO4= plot


- 8 -

The saturation index (SI) is the parameter that indicates the state of saturation of natural waters with respect to relevant minerals. It is defined as SI = log10 (Q/Keq) where Q is the reaction quotient and Keq the equilibrium constant at that temperature. The activities of individual dissolved species were computed by the “PHREEQC” program [14]. The SI values reported in Table 1, considering mineralogical phases in this area such as anhydrite, gypsum, aragonite, calcite, dolomite, magnetite and fluorite, are mainly close to zero or positive, indicating oversaturation or quasi-equilibrium conditions. This agrees with the main lithotypes present in the study area.

Table 1: Saturation indexes for some selected mineralogical phases

SI Date Phase LP G1 G2 G3 Anhydrite -0.22 -0.98 -1.35 -1.10 Aragonite 1.65 0.89 0.75 1.19 Calcite 1.79 1.03 0.89 1.33 Dolomite 4.31 2.42 2.06 3.00 Fluorite -3.96 -2.02 -2.11 -2.15 Magnesite 1.94 0.80 0.61 1.11

Jul. 2000

Gypsum -0.02 -0.79 -1.15 -0.90 Anhydrite 0.15 -0.42 -0.78 -0.60 Aragonite 0.87 0.41 -0.79 0.19 Calcite 1.02 0.55 -0.65 0.34 Dolomite 2.14 1.39 -1.11 1.00 Fluorite -2.12 -2.37 -1.73 -2.44 Magnesite 0.33 0.06 -1.26 -0.16

Apr. 2001

Gypsum 0.39 -0.20 -0.55 -0.38 Anhydrite -0.87 -0.46 -0.93 -0.59 Aragonite 0.57 1.20 1.06 1.20 Calcite 0.72 1.35 1.20 1.34 Dolomite 2.69 3.01 2.66 2.97 Fluorite -1.27 -1.78 -1.81 -1.78 Magnesite 1.20 0.86 0.64 0.84

Oct. 2001

Gypsum -0.65 -0.24 -0.71 -0.37

Comparison between the overall chemical composition of the waters of LP, G1, G2 and G3 lakes during this period of study (Table 2) shows that G1 and G3 have the same chemical composition. On the other hand, comparison of these two with the middle one (G2), highlights the presence of a more diluted water which can be explained through a preferential surface water flow towards the “Gorgo” G2. During the drought period this contribution decreases and the water composition becomes again similar to the others. Comparing the lake LP waters with one of these (for instance G1), shows that it is always more concentrated, even though some linear relationship remains (see Figure 8).


- 9 -

Table 2: Comparison among the chemistry of G1, G2, G3 and LP lakes. Previous chemical analyses are also reported (Consorzio di bonifica “Trapani 1”, 1969-70)

Cond Na K Ca Mg Cl SO4 HCO3Sample mS/cm meq/l meq/l meq/l meq/l meq/l meq/l meq/l LP 1969 11.0 65.2 2.4 2.8 33.2 94.5 28.0 7.6 LP 1970 8.5 79.0 n.d. 3.5 24.7 85.0 13.1 7.9 LP Jul. 2000 16.2 348.8 7.7 36.3 135.0 431.5 137.4 16.1 LP Apr.2001 10.8 87.4 1.2 36.4 43.1 106.5 56.6 5.9 LP Oct. 2001 26.5 226.7 5.0 5.3 68.6 264.1 46.8 12.5 G.1 1969 3.1 16.1 0.9 4.4 7.6 21.9 8.0 5.0 G.1 1970 2.5 21.5 n.d. 3.0 7.2 20.7 2.8 7.2 G.1 Jul. 2000 5.7 43.5 1.7 11.3 17.1 43.3 17.5 6.1 G.1 Apr. 2001 5.8 46.4 1.4 12.7 18.6 54.1 23.9 5.7 G.1 Oct. 2001 5.5 44.7 1.6 12.7 19.2 48.2 21.2 4.6 G.2 1969 1.7 7.4 0.4 4.0 4.8 11.1 12.0 4.2 G.2 1970 1.5 9.6 n.d. 3.4 3.5 10.6 1.5 4.0 G.2 Jul. 2000 3.1 20.3 0.7 6.6 6.1 21.2 7.6 4.6 G.2 Apr. 2001 3.0 22.0 1.9 7.3 9.0 27.9 10.5 4.6 G.2 Oct.2001 3.6 20.1 0.7 6.2 7.9 21.9 7.7 3.5 G.3 1969 3.1 16.1 1.3 4.0 6.0 22.3 5.6 6.8 G.3 1970 2.5 21.0 n.d. 2.9 6.2 21.8 0.8 6.3 G.3 Jul. 2000 4.8 36.1 1.7 9.3 13.2 38.1 14.1 5.9 G.3 Apr. 2001 4.8 39.1 1.8 6.7 14.6 44.8 18.8 3.6 G.3 Oct. 2001 6.0 36.3 1.8 10.1 14.1 41.4 16.1 4.2

july 2000

apr. 2001

oct. 2001

0

100

200

300

400

500

0 10 20 30 40 50 60G1 (meq/l)

LP(m

eq/l) Na

Na

Na

Cl

Cl

ClSO4

SO4SO4

MgMg

MgCaCaCaHCO3K

Figure 8: Comparison between the overall chemistry of LP and G.1 lakes This may be explained with a simplified model reported in Figure 9. According to it, when the LP lake is full part of its water can pour into the other lakes modifying their chemical composition. This model also takes into account the piezometric levels measured along the karst depression. In fact, from the hydrogeological scheme it is possible to note that the lakes water levels in all the samplings evidence water flow from LP towards G1 and, at the same time, from G2 towards G1 and G3. G2 works, in somehow, as a hydrogeological watershed.


- 10 -

Figure 9: Water circulation model between LP and G.1, G.2 and G.3 lakes A comparison of previous chemical analyses carried out in the same area [5,18] underlines an increase in lake water salt concentration over the time. This increase may be due to:

� over exploitation of surrounding wells during the period of major surface water contribution causing a decrease in lake input;

� decrease of average annual rainfall, with a historical minimum in 1999 (Figure 10); � greater evaporation due to the increase of the average annual temperature (Figure 10);

348,0

506,0

245,5

438,6

645,5713,5

456,4433,6

483,9560,9

543,2

470,1444,0

455,0411,2

474,8483,7

406,3

534,9586,9

309,3364,3

555,3

277,7

800,71082,2

21,021,1

21,0

19,219,6

18,818,4

19,218,018,1

18,118,5

18,118,2

18,117,8

18,017,9

17,7

17,517,2

16,7

17,2

16,7

17,618,1

200300400500600700800900

10001100120013001400150016001700

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

year

Rainf

all(m

m)

5678910111213141516171819202122

Temp

eratu

re(°C

)

RainfallTemperature

Figure 10: Precipitation and Temperature values in the last 25 years The δ18O composition of the meteoric waters is reported in Table 3. Annual weighed means resulted –6.3 ‰ (March 2000 - February 2001) and –6.7 ‰ (March 2001 – February 2002).


- 11 -

Table 3: δ18O values for the meteoric waters. (a.w.m. = annual weighed mean; 0= no rain; n.d.= not determined)

Year Month vol.(ml) δ18O (‰) awm(‰) March 300 -4.4 April 1,600 -2.8 May 1,910 -5.7 June 600 1.0 July 250 -1.2

August 0 n.d. September 750 -1.7

October 3,900 -6.0 November 1,750 -3.6

2000

December 5,550 -6.5 January 7,000 -7.9 February 7,512 -7.5

-6,3

March 2,750 -6.5 April 850 -6.6 May n.d. n.d. June 0 n.d. July 0 n.d.

August 0 n.d. September 450 -1.0

October 650 -2.1 November n.d. n.d.

2001

December January 4,550 -8.0 February 575 -7.5

-6,7

March 2002 April 1,900 -4.9

The second mean resulted more negative due to the absence of rains during warmer months, generally characterized by more positive values. Figure 11 shows the isotope composition of all waters analysed in this study.

-7-6-5-4-3-2-10123456

P1 P7 P8 P9 P10

P11

P12

P13

P15

P16

P17

P18 LP G1 G2 G3 P20

P22

P23

P24

P25

P26

P27

P28

P29

P30

P31

P32

P33

P34

P35

P36

P37

P38

Wells and Lakes

δδ δδ18O

‰

july 2000apr. 2001oct. 2001well linemeteoric lineSeawater line

Figure 11: δ18O of the waters sampled. The annual weighed mean for the meteoric water and the seawater values are also reported


- 12 -

Almost all well waters result quite homogenous and close to the local meteoric recharge value (around –5 ‰, which is a little more positive in relation to meteoric value because of surface evaporation before the infiltration) except P12, and lake waters, which exhibit the most positive values. Consequently, δD measurements for these samples were performed (Table 4) and the values have been reported on the δD - δ18O diagram, along with the meteoric water line (M.W.L.) established by J. R. Gat and I. Carmi [11] for the Mediterranean area (Figure 12). The values lie along a line whose slope is typical for an evaporative process where the “Gorghi tondi” water lakes are the highest end-members observed.

Table 4: δD and δ18O values of the lakes. Values are in ‰ vs. V-SMOW July-00 Apr.-01 Oct-01 Sample δD‰ δ18O‰ δD‰ δ18O‰ δD‰ δ18O‰

G1 21 4,5 12 3,0 19 4,5 G2 16 3,7 11 2,8 15 3,8 G3 17 3,4 11 2,3 12 3,5 LP 5 1,4 3 0,6 8 1,3

Seawater 9,8 1,3 n.d. n.d. n.d. n.d.

δD = 4.2δ18O - 0.84

-40

-30

-20

-10

0

10

20

30

40

-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7δδδδ18O‰

δδδδD‰

july 2000apr. 2001oct. 2001SeawaterM.W.L.average lakes

δD = 8δ18O +22

Figure 12: δD vs. δ18O plot relative to lakes values. It also reported the meteoric water line (M.W.L.) defined by Gat and Carmi (1970) for the Mediterranean area. The intercept on this line, considering the mean δ18O value obtained for the meteoric precipitation, gives a δD value around -30 %o, which is typical for the precipitation in this area


- 13 -

CONCLUSIONS In the investigated area the main aquifer present is located in the calcarenites and is fed by local meteoric recharge. Lakes “Preola” and “Gorghi tondi” display salt concentrations higher than the aquifer, which supplies them, suggesting processes such as water-rock interaction, evaporation and seawater mixing. Seawater intrusion, with a west-east direction, seems more evident in lake “Preola” and in well n. 12. Considering geochemical, isotopic, and hydrological aspects, one can suppose a preferential flow path between P12 and LP with respect to P28, which may be considered as a low permeability zone and hence with a limited water exchange. Similarities among the lakes waters may indicate that among them some mixing process takes place, depending on cyclic changes of their hydraulic loads. The increase in salt concentration in the lakes, since 1969 to present day, is also due to other factors such as: a) reduction in average rainfall, b) over exploitation of the surrounding wells.

ACKNOWLEDGMENT The University of Palermo and the WWF Italia-Natural Reserve “Lago Preola e Gorghi tondi” financed this research.


REFERENCES [1] Aruta L. & Buccheri G. (1971). Il Miocene preevaporitico in facies carbonatica-

detritica dei dintorni di Baucina, Ciminna, Ventimiglia di Sicilia, Calatafimi (Sicilia). Riv. Min. Sic., vol. 130-132, 188-194.

[2] Boucart J., (1938). Le marge continentale. Essai sur les Règression et trasgressions marines. Bull. De la Soc. Gèol. Del France, s. 5, Tome VIII, 393-474.

[3] Cipolla F. (1934). Nuovi contributi alla geologia e geografia fisica di Mazara del Vallo e suoi dintorni (Prov. di Trapani). Boll. Soc. di Scienze Naturali ed Economiche di Palermo, anno XV, 1934-XII.

[4] Coleman M. L., Shepherd T. J., Durham J. E., Rouse J. E., Moore G. R. (1982). Reductions of water with zinc for hydrogen isotope analysis. Anal. Chem., 54, 993-995.

[5] Consorzio di bonifica “Trapani 1”, (1969-1970). Relazione idrogeologica per la costruzione di un invaso a scopo irriguo tra Mazara del Vallo e Campobello di Mazara. Volume I, II.

[6] Costa L., Farinha J. C., Zadis G., Mantzavelas A., Fitoka E., Hecker N. & Tomàs Vives P. (1996). Mediterranean Wetland Inventory. MedWet / Istituto da Conservação da Natureza (ICN) / Wetlands International / Greek Biotope / Wetland Centre (EKBY) Publication. Volume III.

[7] Decima A. & Wezel F. C. (1971). Osservazioni sulle evaporiti messiniane della Sicilia centro-meridionale. Riv. Min. Siciliana,12 (130-132), 172-187.

[8] Epstein S., Mayeda T. (1953). Variations of O18 content of waters from natural sources. Geochim. Cosmochim. Acta, 213.

[9] Farinha J. C., Costa L. T., Hecker N. & Tomàs Vives P. (1996). Mediterranean Wetland Inventory. MedWet / Instituto da Conservação da Natureza / Wetlands International Publication.

[10] Forti P., Sauro U. (1997). Gypsum karst of Italy. Gypsum Karst of the World. Int. J. Speleol. 25 (34), 239-250.

[11] Gat J. R., Carmi I. (1970). Evolution of the isotopic composition of atmospheric water in the Mediterranean Sea area. J. Geophys. Res. 75, 3039-3048.

[12] Gonfiantini R., (1978). Standards for stable isotope measurements in natural compounds. Nature, 271, 534-536.

[13] Greenberg A. E., Clesceri L. S. (1992). Standard Methods for examination of water and wastewater. A.D. Eaton Editour., 18th edition.

[14] Parkhurst D. L. (1995). PHREEQ-C, a computer program for speciation, reaction path, advective transport and inverse geochemical calculations. USGS, Water Resources Investigations, Report 95-4227, Lakewood, CO.

[15] Piper A. M.(1944). A graphic procedure in the geochemical interpretation of water analyses. Am. Geophys. Union Trans., v.25, p. 914-923.

[16] Ruggieri G., Unti M., Morani M. A., (1975). La Calcarenite di Marsala (Pleistocene inf.) ed i terreni contemporanei. Boll. Soc. Geol. It., 94,1623-1627.


[17] Tulipano L., (2001). La progressiva salinizzazione delle acque sotterranee nell’acquifero carsico della penisola Salentina. In Parise M. (ed.), Atti Tav. Rot. Acque del Terzo Millennio, Grotte e dintorni, 1-2: 29-38. Grafichena: Fassano.

[18] Vassallo M. (2001). Caratterizzazione geochimica e idrogeologica delle acque della Riserva Naturale “Lago Preola e Gorghi Tondi” (Mazara del Vallo, Sicilia). Tesi di Laurea, A. A. 2000-2001, pp.88, Università degli Studi di Palermo.

[19] Zadis G.C., Mantzavelas A.L. & Fitoka E.N. (1996). Mediterranean Wetland Inventory. MedWet / Greek Biotope / Wetland Centre (EKBY) / Istituto da Conservaçào da Natureza / Wetlands International Publication. Volume IV.

e-water · circulating within the calcarenitic aquifer. however, the lake waters chemical...

Documents